Melanosomes are subcellular organelles in which melanin pigments are synthesized and stored within skin melanocytes and pigment cells of the eye. Within melanosomes, newly synthesized melanins polymerize on a sheet-like matrix of proteinaceous fibers that are largely composed of a single protein, PMEL. PMEL fibrils have biophysical features of amyloids like those associated with neurodegeneration in Alzheimer and Parkinson Diseases, but PMEL fibrils are non-pathogenic and serve a physiological role in pigmentation. By assessing how PMEL fibrils form within melanosome precursors and how their formation differs from that of pathogenic amyloid fibrils, we may better understand how to divert pathological amyloid into non-pathological forms. This proposal aims to (1) define molecular mechanisms that regulate physiological PMEL amyloid formation, and (2) test whether mutations that alter PMEL amyloid properties confer toxicity. Whereas pathological amyloids form slowly in solution, in vitro PMEL rapidly polymerizes into amyloid upon dilution out of denaturant, suggesting that the rate of fibril formation might be critical to distinguish functiona from pathological amyloid. How fibril formation is regulated in vivo is not known. Our preliminary data show that mature forms of PMEL just prior to fibril formation are disulfide bonded to a protein that we are in the midst of identifying. Mutations that interfere with this association hav no effect on early PMEL folding, processing or trafficking to endosomal compartments, but ablate the ability of PMEL to polymerize into amyloid fibrils in cells. This association thus likel impacts conformational changes required for amyloid formation. Preliminary data also show that fibrils begin to form in endosomes in association with intralumenal membrane vesicles (ILVs), suggesting that the ILVs themselves participate in amyloid biogenesis. Here, we will test the role of each of these factors in PMEL amyloid formation, and thereby begin to define the molecular mechanisms underlying the conformational change that converts PMEL from a soluble protein into physiological amyloid. Mice that lack a functional Pmel gene generate melanosomes without fibrils but have only modest defects in skin pigmentation. By contrast, several animal models bearing PMEL orthologues with dominant mutations are severely hypopigmented; human PMEL mutants bearing these mutations generate fibrils that have distinct morphological and functional properties and that inhibit melanogenesis in cultured cells. We hypothesize that the mutations render functional PMEL amyloid toxic to melanocytes. We will test this hypothesis in Aim 3 by expressing the mutant PMEL forms in mice and assessing their effect on pigment cell viability. Our Specific Aims are: 1. To determine how a disulfide bonded associated protein facilitates PMEL amyloid formation; 2. To test whether PMEL amyloid formation is seeded by endosomal membranes; and 3. To test whether dominant mutations in PMEL generate toxic amyloid in pigment cells.